Esm2 Variant Effect Tools

Configuration File: esm2_variant_effect_tools.json Tool Type: Local Tools Count: 1

This page contains all tools defined in the esm2_variant_effect_tools.json configuration file.

Available Tools

ESM2_score_missense_variant (Type: ESM2VariantEffectTool)

Score a missense protein variant with ESM-2 masked-marginal log-likelihood ratio (Meier 2021) ove…

ESM2_score_missense_variant tool specification

Tool Information:

  • Name: ESM2_score_missense_variant

  • Type: ESM2VariantEffectTool

  • Description: Score a missense protein variant with ESM-2 masked-marginal log-likelihood ratio (Meier 2021) over HuggingFace’s keyless hf-inference. Input wild-type 1-letter sequence + 1-based position + mutant residue; returns LLR = logP(mut) - logP(wt). Negative => mutant disfavored (candidate deleterious). No API key. For ESMC/SAE scoring with a key use ESM_score_sequence.

Parameters:

  • sequence (string) (required) Wild-type protein sequence in 1-letter amino-acid code (whitespace is ignored).

  • position (integer) (required) 1-based residue position of the variant within the sequence.

  • mutant (string) (required) Mutant amino acid (single 1-letter code, one of the 20 standard residues).

  • wild_type ([‘string’, ‘null’]) (optional) Optional wild-type residue (1-letter). If given, it is validated against the residue at position to catch coordinate/sequence mismatches.

  • model_id ([‘string’, ‘null’]) (optional) ESM-2 model to use. Default ‘facebook/esm2_t33_650M_UR50D’. Other sizes: esm2_t30_150M_UR50D, esm2_t12_35M_UR50D, esm2_t6_8M_UR50D.

  • wait_for_model ([‘boolean’, ‘null’]) (optional) If true, block until the model finishes loading on the HF servers instead of returning a retryable ‘loading’ status.

Example Usage:

query = {
    "name": "ESM2_score_missense_variant",
    "arguments": {
        "sequence": "example_value",
        "position": 10,
        "mutant": "example_value"
    }
}
result = tu.run(query)